Current fiber-optic long-haul communication networks are predominantly comprised of point-to-point fiber-optic links. Data-modulated optical signals originated at one end propagate through the fiber medium to the opposite end. While propagating, the optical signals suffer attenuation due to scattering in the fiber medium, as well as losses in other components such as connectors. To compensate for the loss, optical amplifiers are placed at regular intervals along the fiber span, typically 40 to 100 km apart. A single fiber strand can carry independent multiple (e.g. up to 100 or more) optical signals, each signal being differentiated by a slightly different wavelength. Thus, optical amplifiers amplify all the wavelengths simultaneously. For several reasons, it is common for the optical amplifiers to be operated in a saturated regime with a fixed total optical power output, but variable gain.
Optical communication networks have, however, begun to evolve away from simple point-to-point links. The first step was the introduction of fixed optical add-drop multiplexers (OADMs). The OADMs are positioned at intermediate points along the fiber-optic link between the terminal ends, and provide the capability of adding or dropping individual wavelengths. This diversity of signal origination and termination points allows for more flexible and useful optical network architectures. Another step in the evolution has been the addition of dynamic OADM capability. Dynamic OADMs facilitate dynamic switching and rerouting of individual optical wavelength signals between various fiber-optic links.
Switching and rerouting of individual optical wavelengths between fiber-optic links, however, creates difficulties with respect to controlling optical power in each wavelength. As discussed above, optical amplifiers are commonly operated such that they provide a fixed total output power, which is then shared among the various wavelengths. This provides an undesirable coupling mechanism among the optical wavelengths. Optical wavelength signals can appear and disappear in the fiber-optic link, either due to component failures and/or fiber cuts in the fixed OADM case, or due to active wavelength switching in the dynamic OADM case. As optical wavelength signals disappear, optical amplifiers allocate the unused power to the remaining signals, potentially causing a substantial increase in their power. Conversely, newly added optical wavelengths can cause substantial power drop in the existing wavelengths.
These optical power transients can be detrimental for several reasons: 1) optical power levels exceeding receiver dynamic range can cause loss of data on the low end and potential permanent component damage on the high end; 2) reduced optical wavelength power can cause signal to noise degradation and may result in a loss of data; 3) increased optical wavelength power can cause nonlinear signal distortions and noise and may result in a loss of data; and 4) optical power transients may disrupt seemingly unrelated parts of the network, complicating alarm management and troubleshooting.
Several approaches to solving this problem have been proposed. The majority of the proposed solutions have concentrated on controlling the gain of optical amplifiers to keep it substantially constant and independent of the number of channels. While feasible, these approaches suffer from several fundamental drawbacks. For example, these approaches require modification or replacement of fielded optical amplifiers, resulting in significant cost increase for the optical network. Typical approaches also waste optical pump power, thereby reducing useful optical power available from the amplifier. In turn, this may limit the optical network capacity and/or span designs. Moreover, operation of optical amplifiers in a “Constant gain” mode precludes self-healing of amplifier chains, whereby degradation in one optical amplifier gain or fiber span loss is compensated for by an automatic increase in gain by subsequent amplifiers.
According to another approach to controlling or preventing optical power transients, a single compensating optical wavelength is added to an aggregate signal containing of the independent signals to be transmitted on a fiber link. The total power propagating in the link is measured and kept constant by adding or removing power at a predetermined optical wavelength. ,This approach overcomes most of the disadvantages associated with amplifier gain control methods, but introduces other difficulties. In particular, having only one compensating wavelength may not be sufficient for stabilizing channels that are spectrally far removed. Also, networks comprised of multiple cascaded OADMs may introduce difficulty in selecting proper wavelength for the compensating channel, and may necessitate many stages of sequential transient control, progressively reducing its effectiveness.
Accordingly, there is a need in the art for a system and method for optical power transient control and prevention in communication networks that overcomes the disadvantages of the prior art.